- From: David Hyatt <hyatt@apple.com>
- Date: Thu, 21 Oct 2010 15:00:45 -0500
- To: "L. David Baron" <dbaron@dbaron.org>
- Cc: www-style@w3.org
On Oct 21, 2010, at 2:35 PM, L. David Baron wrote: > I wanted to raise a third option for how to deal with logical > properties in addition to the two I cited in > http://lists.w3.org/Archives/Public/www-style/2010Jun/0003.html . > > The proposal now in > http://dev.w3.org/csswg/css3-writing-modes/#logical-prop requires > strict adherence to CSS cascading rules. An implementation can do > this using the approach described in > http://lists.w3.org/Archives/Public/www-style/2002Sep/0049.html . > > However, if we allow implementation-defined deviation from the CSS > cascading rules when determining the relative priority of > declarations of a logical and physical property **in the same > ruleset**, then I think the required implementation complexity is > significantly reduced. (If we did this, I would suggest making use > of such property pairs within the same ruleset an authoring > conformance error.) > > In particular, with this approach, implementations would no longer > need the extra pseudo-properties to track the relative priority of > declarations within the same declaration, and they could do the > relevant computations at cascade time after determining the > direction. > > (Getting this reduction in complexity might somewhat limit > implementation approaches, since it requires computing 'direction' > before cascading any properties with logical/physical pairs. I > think that's not a problem in Gecko, so this approach would simplify > our implementation of logical properties significantly. I'd be > interested to hear if it would help in other implementations.) Can you explain this further? I don't really understand what you're suggesting. Examples would help. In WebKit, which implements all of the logical properties, we just resolve direction and writing-mode first (similar to what implementations have to do for font), and then we map the logical property to a physical property. This is all very simple and takes very little code. Is there some problem with this approach that I'm missing? dave (hyatt@apple.com)
Received on Thursday, 21 October 2010 20:01:19 UTC